You are on page 1of 4

Astron. Astrophys.

331, L33–L36 (1998)


ASTRONOMY

LETTER
AND
ASTROPHYSICS
Letter to the Editor
The adoption of ICRS on 1 January 1998:
meaning and consequences
M. Feissel1 and F. Mignard2
1
IERS Central Bureau, Observatoire de Paris, URA CNRS 1125, 61 av. de l’Observatoire, F-75014 Paris, France
2
Observatoire de la Côte d’Azur, CERGA, UMR CNRS 6527, av. Copernic, F-06130 Grasse, France

Received 28 November 1997 / Accepted 9 December 1997

Abstract. As of 1 January 1998, the celestial reference system system. During the 1994-1997 term, the WGRF concluded this
adopted by the IAU will be based on a set of extragalactic di- preparatory phase by recommending that the 1997 IAU General
rections, replacing the Fundamental Katalog No 5 (FK5) that Assembly adopt the IERS celestial reference system under the
gave star coordinates in a reference system (referred to as equa- name International Celestial Reference System (ICRS) and to
tor and equinox J2000) based primarily on the dynamics of the charge the IERS to maintain its realization, the ICRF (Interna-
solar system. This change is deeper than a mere replacement of tional Celestial Reference Frame), in cooperation with an IAU
a reference system by a better one. It implies a drastic modifica- working group on reference frames.
tion in the underlying concepts and has practical consequences The fixed directions underlying the FK5 (Fricke et al. 1988)
at various levels for the users needing the best astrometric ac- and the preceding fundamental star catalogues of positions and
curacy. The IAU decision begins a new era in the definition of proper motions were defined by the mean equator and equinox
reference systems and marks a sharp break with the practice of at a reference epoch (J2000.0 for the FK5, B1950 for the FK4),
the last three hundred years characterized by the long series of which implied dynamical modelling of the Earth’s orbital mo-
fundamental star catalogues. tion (ecliptic, equinox) for the realization of the reference frame
and its subsequent access. In the past, these fundamental di-
rections were considered conventional for some decades; they
1. Historical background were changed from time to time, in particular to take advan-
tage of advances in modelling of the motion of solar system
1.1. The quasi-inertial frame objects. As was originally proposed by Guinot (1979), the IAU,
Adopting extragalactic directions as primary celestial references in its 1991 Recommendations on Reference Systems (Bergeron
instead of star positions has been envisioned for over a decade 1992), decided to select distant extragalactic objects as the ba-
by the International Astronomical Union (IAU). A series of sis of its new celestial reference system and to adopt directions
Working Groups was organized jointly by several IAU com- that would remain fixed with respect to a selected set of these
missions mainly to establish lists of objects appropriate for ob- objects. Continuity between the old and new system was manda-
servation by Very Long Baseline radio Interferometry (VLBI) tory so that these directions would be consistent with their pre-
at wavelengths of 13 and 3.6 cm. VLBI is the technique that vious realizations, i.e., the pole and origin of right ascensions
measures most efficiently directions of compact extragalactic in the new reference system should be close to the FK5 pole
radio sources, reaching sub-milliarsecond (mas) accuracy. and right ascension origin, within the uncertainties of the FK5.
A decisive breakthrough was the formulation by the Work- A fundamental advantage of selecting extragalactic objects is
ing Group on Reference Systems (1988-1991) of a comprehen- that they are so distant that their proper motions are currently
sive group of recommendations that dealt with both the funda- undetectable, even by the most precise techniques. More quan-
mental concepts of space-time references and their actual real- titatively, if we assume that the tangential velocity of a distant
izations. During the 1991-1994 term, the IAU Working Group extragalactic object is comparable to its recession speed, the an-
on Reference Frames (WGRF) produced a list of about 600 radio gular motion would be given by the Hubble constant, or about
sources that could be used to realize an extragalactic celestial 0.01 mas/a.
reference system, provided that careful observations could tie According to the new rules adopted in the 1991 IAU Rec-
them together into a consistent system. Meanwhile the Interna- ommendations, the fundamental directions of the celestial refer-
tional Earth Rotation Service (IERS) had initiated and devel- ence system will remain fixed in space, as long as the directions
oped operational procedures to maintain the celestial reference of the distant extragalactic sources can be regarded as fixed di-
rections. They will no longer be dependent on precise modelling
Send offprint requests to: M. Feissel (e-mail: feissel@obspm.fr)
L34 M. Feissel & F. Mignard: The adoption of ICRS on 1 January 1998:meaning and consequences
LETTER

of the motion of solar system objects and on the definition of pole is in the J2000.0 direction defined by the conventional IAU
the equinox by means of the celestial equator and the ecliptic. models for precession (Lieske et al. 1977) and nutation (Sei-
The adopted positions of the defining sources may be re- delmann 1982). The origin of right ascensions is also defined
estimated when improved information is available, but the di- consistently with that of the FK5 by fixing the right ascension
rection of the coordinate axes will be maintained by implement- of 3C 273B to the Hazard et al. (1971) FK4 value transformed
ing the statistical condition that the new coordinates of selected to the J2000(FK5) System (Kaplan et al. 1982).
sources show no global rotation with respect to the old set. It The Hipparcos star positions and proper motions (ESA
is also foreseen that some sources may be deleted or new ones 1997) and the JPL solar system ephemerides (starting with
could be added in the future. This will be done by adhering DE403, see Standish 1997) are already expressed in the ICRS
strictly to a predefined procedure. and several new catalogues (Urban et al. 1997a,b) in the visible
In order that the use of this new reference system should not have been aligned with the Hipparcos system, the latter being
be detrimental to the analysis of the highly accurate astrometric used as a secondary realization of the ICRS. The International
techniques, the IAU explicitly introduced the Theory of General Terrestrial Reference System (ITRS), also maintained and made
Relativity as the background for all theoretical and data analysis available by the IERS as the International Terrestrial Reference
problems related to time and space. Frame (ITRF), is connected to the ICRS at any time by the IERS
Earth Orientation Parameters (EOP) with an accuracy equiva-
lent to 1 cm at the surface of the Earth.
1.2. The IAU Resolutions
At its 23rd General Assembly in August 1997, the IAU decided
2.2. ICRF: The International Celestial Reference Frame
(Appenzeller 1998),
The choice of extragalactic objects to realize the fiducial di-
– That, as from 1 January 1998, the IAU celestial reference rections was made possible by the availability of a mature and
system shall be the International Celestial Reference System highly precise observing technique, Very Long Baseline radio
(ICRS) as specified in the 1991 IAU Resolution on Ref- Interferometry (VLBI). A detailed description of astrometric
erence Systems and as defined by the International Earth and geodetic VLBI analysis is given by Sovers and Jacobs
Rotation Service (IERS); (1996). The major sets of software in use for this modelling have
– That the corresponding fundamental reference frame shall been intensively tested and compared to each other. They are
be the International Celestial Reference Frame (ICRF) con- considered to be implemented consistently within one picosec-
structed by the IAU Working Group on Reference Frames; ond for the delay and one femtosecond/second for the delay rate,
– That the Hipparcos Catalogue shall be the primary realiza- one order of magnitude better than the observations (Jacobs et
tion of the ICRS at optical wavelengths; al. 1997).
– That IERS should take appropriate measures, in conjunc- The ICRF consists of a catalogue of equatorial coordinates
tion with the IAU Working Group on Reference Frames, to of 608 extragalactic radio sources derived from about 1.6 million
maintain the ICRF and its ties to the reference frames at observations accumulated by a worldwide network over 1979-
other wavelengths. 1995. It was derived by a sub-group of the WGRF that agreed
In the next two sections we briefly describe the ICRS, the on an optimal data analysis strategy (Ma et al. 1998). It includes
ICRF and the Hipparcos catalogue, summarize the basic prin- three lists of objects, with
ciples of the maintenance of the ICRF, and list the main conse- 1. the most compact and best observed 212 defining sources,
quences of the change. with a median uncertainty of 0.4 mas on individual positions,
2. compact sources (294) whose positions are likely to be im-
2. ICRS and ICRF, Hipparcos, FK5, Equator and Equinox proved when more observations are accumulated in the fu-
ture,
By Reference System it is meant the set of prescriptions and 3. sources less appropriate for astrometric purposes (102), but
conventions together with the modelling required to define, at which are provided for possible ties of reference frames at
any time, a triad of axes. other wavelengths or for other objectives.
By Reference Frame it is meant a practical realization with
given fiducial directions agreeing with the concepts introduced The accuracy of the ICRF realization of the ICRS axes is esti-
in the corresponding Reference System. mated to be 0.02 mas.
Extensive information on the ICRF is available in (Ma &
Feissel 1997) and electronically (anonymous ftp and WWW
2.1. ICRS: The International Celestial Reference System
site: hpiers.obspm.fr).
The ICRS (Arias et al. 1995) complies with the conditions speci-
fied by the 1991 IAU Recommendations. The origin of the ICRS
2.3. The Hipparcos Catalogue and the ICRF
axes is located at the barycentre of the solar system and the di-
rections of its axes are fixed relative to distant extragalactic The direct result of the data processing of the Hipparcos observa-
sources. For the sake of continuity with the FK5 pole, the ICRS tions was a catalogue of astrometric positions for about 120 000
M. Feissel & F. Mignard: The adoption of ICRS on 1 January 1998:meaning and consequences L35

LETTER
Table 1. Global orientation and spin between the FK5 and the ICRS at Table 2. ICRS coordinates of the mean celestial pole (x , y ) and of
J2000.0. The Hipparcos Catalogue has been used as intermediate frame the equinox (z ), see Fig. 1.
and the uncertainties propagated accordingly (Mignard & Froeschlé
Orientation (mas)
1997).
J2000.0 σ
Orientation (mas) Spin(mas/a)
J2000.0 σ σ x : - 5.1 0.01
y : 17.2 0.01
x : -19.9 2.3 ωx : -0.30 0.27 z : -78 10
y : - 9.1 2.3 ωy : 0.60 0.27
z : 22.9 2.3 ωz : 0.70 0.27

PJ2000
•5.1 mas
stars of exceptionally good precision and free of regional er-
ror at the level of 0.1 mas. The intermediate frame in which
these positions and proper motions were given was still arbi- 17.2 mas
trary. A connection between the intermediate Hipparcos frame
and the ICRS was carried out by using various data and methods PICRS
• α = 6h
(Lindegren & Kovalevsky 1975; Kovalevsky et al. 1997). The
resulting alignment of the published Hipparcos Catalogue with 9.1 mas
the ICRS was eventually realized with standard errors for each
axis of 0.6 mas for the orientation at J1991.25 and 0.25 mas/a PFK5 • 19.9 mas
for the spin. Consequently the Hipparcos Catalogue provides
the master realization of the ICRS in the optical domain with
the above uncertainties. α = 0h

22.9 mas 78 mas


2.4. The FK5 and the ICRS • • • α
Ο FK5 Ο ICRS γ J2000
As the complete FK5 catalogue was observed by Hipparcos,
the directions of the pole and origin of right ascensions of the Fig. 1. Location of the poles and the origin of right ascensions of the
FK5 relative to Hipparcos were determined, with an accuracy FK5 and that of the celestial pole and equinox at J2000.0 with respect
of 2.3 mas at the mean epoch J1991.25. The relative spin be- to the ICRS (values of Tables 1 & 2).
tween the FK5 and Hipparcos frames was also assessed from
the systematic deviations between the proper motions, with an
accuracy of 0.1 mas/a. Eventually the directions of the FK5 DE200, Folkner et al. (1994) derived the offset of the mean
axes in the ICRS were estimated (Table 1), taking into account equinox at J2000.0 relative to the ICRS origin of right ascen-
the propagation of the uncertainty of the Hipparcos tie to the sions. Resulting values are summarized in Table 2.
ICRS. The modelling of the FK5 non global deformations af- The infinitesimal rotations of Tables 1 and 2 can be trans-
fect marginally the estimation of the orientation and spin. formed into the coordinates of the poles on a plane tangent to
The infinitesimal angles x , y , z are defined as direct ro- the ICRS pole, as shown in Fig. 1. The position of the origins
tations around the ICRS axes pointing in the directions α = 0 of right ascensions are also depicted along the equator.
h, α = 6 h, δ = 90◦ respectively; ωx , ωy , ωz are defined as their
corresponding time derivatives.
3. Practical consequences of the adoption of the
2.5. The mean pole and equinox at J2000.0 in the ICRS ICRS/ICRF

According to their authors, the realizations of the mean pole and Kovalevsky & McCarthy (1997) have considered the various
equinox at J2000.0 by means of the FK5 (Fricke et al. 1988) consequences of the change from the J2000 Reference System
have respective uncertainties of 50 mas and 80 mas. VLBI and to ICRS, with particular emphasis on the conceptual aspects.
Lunar Laser Ranging (LLR) allow more accurate determination We summarize hereafter the consequences in every-day life of
of these physical directions in the ICRS. the astronomers.
Using a state of the art precession-nutation model (Mc-
Carthy 1996), the analysis of long VLBI series of the observed – For all uses with accuracy requirement less stringent than
motion of the celestial pole in the ICRS allows us to derive 50 mas, the adoption of ICRS has no significant effect.
the coordinates of the mean pole at J2000.0 in the ICRS (IERS – There is no epoch attached to the system, i.e., future updates
1997). Combining VLBI and LLR Earth orientation and terres- of the ICRF will not change the ICRS origin and that of its
trial reference frames and using the JPL planetary ephemeris defining directions.
L36 M. Feissel & F. Mignard: The adoption of ICRS on 1 January 1998:meaning and consequences
LETTER

– Changes of stellar positions between two epochs are derived Jacobs C.S., Sovers O.J., Gordon D., Ma C., Gontier A.-M., 1997, The
by allowing for proper motion. The new positions are still accuracy of the ICRF: an intercomparison of VLBI analysis soft-
referred to the ICRS. ware, JD7, 23rd IAU General Assembly (August 1997). Highlights
– For precise applications, the IERS 1996 precession-nutation in Astronomy.
model (McCarthy 1996) or the celestial pole offsets pub- Kaplan G.H., Josties F.J., Angerhofer P.E., Johnston K.J., Spencer J.H.,
1982, AJ, 87, 570
lished by the IERS should be used, rather than the less accu-
Kovalevsky J., McCarthy D., 1997, Astronomical effects of current
rate IAU conventional models. The IERS 1996 model being
changes in fundamental astrometric references, JD3, 23rd IAU
referred to the mean pole at J2000.0 that is offset from the General Assembly (August 1997). Highlights in Astronomy.
ICRS one, two additional constants must be used in order to Kovalevsky J., Lindegren L., Perryman M., 1997, A&A, in press.
refer the motion to the ICRS pole. Estimates of these con- Lieske J.H., Lederle T., Fricke W., Morando B., 1977, A&A, 58, 1.
stants corresponding to Fig. 1 are - 43.1 mas in longitude Lindegren L., Kovalevsky J., 1995, A&A, 304, 189.
(-17.2 mas/sin ) and - 5.1 mas in obliquity (IERS 1997). Ma C., Feissel M., 1997, Definition and Realization of the International
– The direction of celestial objects in the ICRS are consistent Celestial Reference System by VLBI Astrometry of Extragalactic
with the terrestrial coordinates in the ITRS by the use of the Objects, IERS Technical Note No 23, Observatoire de Paris.
IERS Earth orientation parameters (universal time, polar Ma C., Arias E.F., Eubanks T.M, Fey A.L, Gontier A.-M, Jacobs C.S.,
motion, precession-nutation). Sovers O.J., Archinal B.A., Charlot P., 1998, The International
Celestial Reference Frame as realized by Very Long Baseline In-
The IAU Division 1 (Fundamental Astronomy) has set up an terferometry, AJ, submitted.
McCarthy D.D.(ed.), 1996, IERS Conventions (1996), IERS Technical
ICRS Working Group, in charge of studying all consequences of
Note No 21, Observatoire de Paris.
the adoption of the ICRS, and proposing solutions. This Work- Mignard F., Froeschlé M., 1997, Linking the FK5 to the ICRF, JD12,
ing Group is chaired by François Mignard. 23rd IAU General Assembly (August 1997). Highlights in Astron-
omy.
References Seidelmann P.K., 1982, 1980 IAU Nutation: Celest. Mech., 27, 79.
Sovers O.J., Jacobs C.S., 1996, Observation model and parameter par-
Appenzeller I. (ed.), 1998, Proceedings of the 23rd General Assembly, tials for the JPL VLBI parameter estimation software MODEST,
Transactions of the IAU 23B. JPL Publ. 83-39, Rev. 6, Jet Propulsion Laboratory, Pasadena, Cal-
Arias E. F., Charlot P., Feissel M., Lestrade J.-F., 1995, A&A, 303, ifornia.
604. Standish E.M., 1997, Linking the Dynamical Reference Frame to the
Bergeron J. (ed.), 1992, Proceedings of the 21st General Assembly, ICRF, JD7, 23rd IAU General Assembly (August 1997). Highlights
Transactions of the IAU 21B, 41. in Astronomy.
ESA, 1997, The Hipparcos and Tycho Catalogues, ESA SP-1200. Urban S.E., Corbin T.E., Wycoff G.L., 1997a, The ACT Reference
Folkner W.M., Charlot P., Finger M.H., et al., 1994, A&A, 287, 279. Catalog, AJ, in preparation.
Fricke W., Schwan H., Lederle T., 1988, Fifth Fundamental Catalogue Urban S.E., Corbin T.E., Wycoff G.L., Martin J.C., Jackson
(FK5), Part I. Veroff. Astron. Rechen Inst. 32, Heidelberg. E.S., Zacharias M.I., Hall D.M., 1997b, The AC 2000: The As-
Guinot B., 1979, Basic problems in the kinematics of the rotation of the trographic Catalogue on the Hipparcos System, AJ, in preparation.
Earth, Proc. IAU Symp. 82, Time and the Earth’s Rotation, D.D.
McCarthy, J.D.H. Pilkington eds., D. Reidel - Dordrecht, 7.
Hazard C., Sutton J., Argue A.N., Kenworthy C.N., Morrison L.V.,
Murray C.A., 1971, Nature Phys. Sci., 233, 89. This article was processed by the author using Springer-Verlag LATEX
IERS, 1997, 1996 Annual Report, Observatoire de Paris, II-38, II-73. A&A style file L-AA version 3.

You might also like